~aod-build
Skill95 starsChanged 38 days agoApache-2.0View on GitHub
Generate standardized checkpoint reports for multi-phase implementation projects. Use this skill when pausing implementation at strategic milestones (phase completion, user story completion, critical features) to create comprehensive progress reports with task breakdowns, metrics, knowledge base entries, and resume instructions.
In this repository30 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- ~aod-blueprintSkill
- ~aod-bugfixSkill
- ~aod-defineSkill
- ~aod-deliverSkill
- ~aod-discoverSkill
- aod-foundationSkill
- ~aod-kickstartSkill
- aod-lensSkill
- aod-orchestrateSkill
- ~aod-planSkill
- ~aod-project-planSkill
- ~aod-runSkill
- ~aod-scoreSkill
- ~aod-specSkill
- aod-stackSkill
- ~aod-statusSkill
- code-execution-helperSkill
- git-workflow-helperSkill
- kb-createSkill
- kb-querySkill
- root-cause-analyzerSkill
- securitySkill
- tachi-control-analysisSkill
- tachi-infographicsSkill
- tachi-orchestrationSkill
- tachi-report-assemblySkill
- tachi-risk-scoringSkill
- tachi-sharedSkill
- tachi-threat-reportingSkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 8 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈8
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
Create a checkpoint at:
- Distinctive: line 16
Complete phase finish: - All tasks in a phase marked complete
- Distinctive: line 94
Every checkpoint report includes:
- Distinctive: line 111
Task range: Start and end task IDs
- Distinctive: line 118
When prompted for knowledge base entries, provide learnings that:
- Very specific: line 206
Update CHECKPOINTS_README.md with new checkpoint.
- Distinctive: line 306
Create checkpoints proactively: - Don't wait until pausing
- Distinctive: line 307
Capture knowledge fresh: - Document learnings immediately
- Distinctive: line 310
Test resume commands: - Verify they work before distributing
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Implementation Checkpoint Skill
Purpose
Automatically generate standardized checkpoint reports that document implementation progress at strategic milestones. Each report includes completed tasks, remaining work, technical achievements, knowledge base entries, quality metrics, and resume instructions.
When to Use
Create a checkpoint at:
- Complete phase finish - All tasks in a phase marked complete
- User story completion - Entire user story implemented and tested
- Critical milestone - Core functionality ready (e.g., MVP core)
- Before pausing - Stopping for review, testing, or user feedback
- After architect review - Major approval/sign-off received
How to Use
Quick Usage
# Navigate to skill directory
cd .claude/skills/~aod-build
# Generate checkpoint (interactive prompts will follow)
python scripts/generate_checkpoint.py \
--tasks-file ../../specs/{feature-id}/tasks.md \
--output-dir ../../specs/{feature-id} \
--description "{Phase-Description}"Step-by-Step Process
- Analyze completion status
# Script automatically: # - Reads tasks.md # - Counts [X] completed vs [ ] pending # - Calculates progress percentage # - Identifies completed phases # - Determines task range - Provide context when prompted
Key achievements (3-5 bullet points): > - Achievement 1 > - Achievement 2 > - Achievement 3 Knowledge base entries captured (count or 'none'): > 2 [For each KB entry, provide: Title, Problem, Solution] - Review generated report
# Output: CHECKPOINT-{NN}_{Description}_{TaskRange}.md # Location: specs/{feature-id}/ - Verify index updated
# Check: CHECKPOINTS_README.md includes new checkpoint
Checkpoint Naming Convention
Format: CHECKPOINT-{NN}_{Phase-Description}_{TaskRange}.md
Components:
{NN}: Sequential checkpoint number (01, 02, 03...) - auto-detected{Phase-Description}: Milestone name (e.g., "Phase1-2-Complete", "US1-Complete"){TaskRange}: Task IDs covered (e.g., "T001-T030")
Examples:
✓ CHECKPOINT-01_Phase1-2-Complete_T001-T030.md
✓ CHECKPOINT-02_US1-Complete_T001-T039.md
✓ CHECKPOINT-03_MVP-Ready_T001-T101.md
✓ CHECKPOINT-04_ProductionReady_T001-T120.mdSee references/naming_conventions.md for detailed rules.
Required Sections
Every checkpoint report includes:
- Executive Summary - What was completed, what's next
- Completion Status - Progress table with percentages
- What Was Implemented - Detailed task breakdown by phase
- Key Technical Achievements - Highlights and innovations
- Files Created/Modified - Complete file inventory with counts
- Knowledge Base Captured - Critical learnings documented
- Server/Database Status - Current state of running services
- Quality Metrics - Code quality, reviews, test coverage
- Remaining Work - What's left and estimated time
- Resume Instructions - How to continue implementation
Metrics Calculated Automatically
- Progress percentage:
(completed / total) * 100 - Phase completion: Per-phase progress tracking
- Task range: Start and end task IDs
- Next task: First pending task to work on
See references/metrics_formulas.md for formulas and examples.
Knowledge Base Integration
When prompted for knowledge base entries, provide learnings that:
✅ DO include:
- Solved problems that took >30 minutes
- Non-obvious gotchas or edge cases
- Architectural decisions with trade-offs
- Patterns that should be reused
- Issues that could happen again
❌ DON'T include:
- Obvious facts ("we use PostgreSQL")
- Generic statements ("testing is important")
- Implementation progress ("completed task T026")
- Temporary workarounds
Output Files
The skill generates/updates:
- Checkpoint Report:
CHECKPOINT-{NN}_{Description}_{TaskRange}.md- Location:
specs/{feature-id}/ - Size: ~15-20 KB
- Format: Markdown with tables and code blocks
- Location:
- Checkpoints Index:
CHECKPOINTS_README.md(updated)- Location:
specs/{feature-id}/ - Adds new checkpoint entry
- Updates progress tracking table
- Location:
Resume Commands Generated
The checkpoint report includes auto-generated resume commands:
# For team-lead orchestration
/aod.build --resume
# For direct continuation
# Continue from task {NEXT_TASK}Scripts Reference
analyze_tasks.py
Parse tasks.md and extract metrics.
Input: Path to tasks.md file Output: JSON with completion metrics
Usage:
python scripts/analyze_tasks.py specs/001-{{PROJECT_NAME}}/tasks.mdOutput Format:
{
"total_tasks": 120,
"completed_tasks": 32,
"progress_percentage": 26.67,
"phases": [...],
"task_range": "T001-T030",
"next_task": "T031"
}generate_checkpoint.py
Generate checkpoint report from template.
Usage:
python scripts/generate_checkpoint.py \
--tasks-file specs/{feature-id}/tasks.md \
--output-dir specs/{feature-id} \
--description "{Phase-Description}"Interactive Prompts:
- Key achievements summary
- Knowledge base entries (optional)
- Per-entry: title, problem, solution
Output: Complete checkpoint report file
update_index.py
Update CHECKPOINTS_README.md with new checkpoint.
Usage:
python scripts/update_index.py \
--checkpoints-readme specs/{feature-id}/CHECKPOINTS_README.md \
--checkpoint-file CHECKPOINT-{NN}_{Description}_{TaskRange}.md \
--checkpoint-num {NN}Changes Made:
- Adds checkpoint to "Active Checkpoints" section
- Updates progress tracking table
- Updates "Next Checkpoint Planned" footer
Templates Reference
checkpoint_template.md
Complete Markdown template for checkpoint reports. Contains all required sections with template variables for substitution.
Location: references/checkpoint_template.md
Key Variables:
{CHECKPOINT_NUM},{PHASE_DESCRIPTION},{TASK_RANGE}{DATE},{FEATURE_ID},{TOTAL_TASKS},{COMPLETED_TASKS}{PROGRESS_PERCENTAGE},{PHASES_TABLE},{KEY_ACHIEVEMENTS}{KB_ENTRIES},{NEXT_TASK},{RESUME_COMMAND}
CHECKPOINTS_README_template.md
Template for creating the checkpoints index file (first-time use).
Location: assets/CHECKPOINTS_README_template.md
Usage: Copy to specs/{feature-id}/CHECKPOINTS_README.md when creating first checkpoint.
Examples
Example 1: After Completing Phase 1-2
$ python scripts/generate_checkpoint.py \
--tasks-file specs/001-{{PROJECT_NAME}}/tasks.md \
--output-dir specs/001-{{PROJECT_NAME}} \
--description "Phase1-2-Complete"
Analyzing tasks.md...
✓ Found 120 total tasks
✓ Found 32 completed tasks (27%)
✓ Task range: T001-T030
✓ Checkpoint #01 (auto-detected)
Key achievements (3-5 bullet points):
> - Database schema with 6 entities, 19 indexes
> - Architect review APPROVED
> - MVP core: Atomic locking with row-level locks
Knowledge base entries: 5
[... interactive prompts for KB entries ...]
✓ Generated: CHECKPOINT-01_Phase1-2-Complete_T001-T030.md
✓ Updated: CHECKPOINTS_README.md
Next: Resume with /aod.build --resumeExample 2: After Completing User Story 1
$ python scripts/generate_checkpoint.py \
--tasks-file specs/001-{{PROJECT_NAME}}/tasks.md \
--output-dir specs/001-{{PROJECT_NAME}} \
--description "US1-Complete"
✓ Checkpoint #02: US1-Complete (T001-T039)
✓ Progress: 33% (39/120 tasks)
[... prompts and generation ...]
✓ Generated: CHECKPOINT-02_US1-Complete_T001-T039.mdTroubleshooting
Q: Script can't find tasks.md A: Provide absolute path or verify relative path from skill directory
Q: Checkpoint number is wrong
A: Manually specify with --checkpoint-num NN flag
Q: Template variables not substituted
A: Check template file exists in references/checkpoint_template.md
Q: Index file not updating A: Verify CHECKPOINTS_README.md exists in output directory
Best Practices
- Create checkpoints proactively - Don't wait until pausing
- Capture knowledge fresh - Document learnings immediately
- Be specific in achievements - "Implemented X" not "Made progress"
- Validate metrics - Review auto-calculated percentages
- Test resume commands - Verify they work before distributing
Skill Version: 1.0.0 Last Updated: 2025-11-20 Maintainer: Team Lead Agent
Common Rationalizations
| Rationalization | Reality |
| "I'll skip the checkpoint — I'm only mid-phase" | Step 1 (line 17) lists critical milestones too; mid-phase pause without report leaves no resume artifact. |
| "KB entries can wait until after delivery" | Line 307 directs capturing knowledge fresh; deferred entries miss the gotchas described in lines 121-126. |
| "Achievements like 'made progress' are fine" | Line 308 requires specific achievements; vague entries violate the KB DON'T list (line 130). |
| "I can hand-write achievements without the script" | Quick Usage (line 27) requires generate_checkpoint.py — manual reports skip the auto progress percentage from line 109. |
| "Resume commands are documentation, not testing" | Line 311 requires testing resume commands before distributing; untested commands break next session. |
Red Flags
- Agent creates a checkpoint without running
analyze_tasks.pyper the metrics block (lines 162-183). - Agent's KB entries match the DON'T list (line 128): "obvious facts," "generic statements," "implementation progress."
- Agent skips line 310 (validate metrics): publishes auto-calculated percentages without sanity-checking them.
- Agent omits the Required Sections list (line 96) — Quality Metrics, Resume Instructions, or Files Created.
- Agent uses arbitrary checkpoint numbering instead of the auto-detected
{NN}from the script (line 78). - Agent's report contains template variables like
{NEXT_TASK}unsubstituted — Troubleshooting Q3 (line 298) flags this.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

